Pencil Puzzle Bench - Slitherlink — leaderboard

Metric: Direct-ask Success Rate (%). Source: ppbench.com. 51 models tracked.

Top models

#ModelScore
1Claude Opus 4.6 (Thinking)33.3
2GPT-5.2 (High)33.3
3GPT-5.2 (xHigh)20
4Gemini 3.1 Pro (Preview)13.3
5Grok 4.1 Fast (Reasoning)13.3
6Gemini 3 Pro13.3
7GPT-5.2 (Medium)13.3
8GPT-5.2 Pro13.3
9Gemini 3 Pro (High)13.3
10Claude Sonnet 4.6 (Thinking)13.3
11Claude Sonnet 4.5 (Thinking)6.7
12Grok 4.1 Fast6.7
13DeepSeek V3.2 Speciale6.7
14Claude Opus 4.5 (Thinking)6.7
15GPT-5.1 (Medium)6.7

Interactive version: theaggregate.ai/benchmark?slug=pencil-puzzle-bench-slitherlink · How the rankings work · Data refreshed daily, snapshot 2026-07-22.